Tuberculosis (TB) is one of the most important killers of all time. First-line therapy, which consists of combination of rifampin, isoniazid, and pyrazinamide, is under threat from the development of multi-drug resistant TB (MDR-TB) and extensively drug resistant TB. The drug resistance is driven, to a great extent, by poor compliance. It is hypothesized that resistance emerges due to the pharmacokinetic mismatch of component drugs, effective monotherapy of bacilli under acidic and hypoxic conditions, differential post-antibiotic effects of drugs, and prolonged time in the mutant selection window (TMSW). We will test these hypotheses using pharmacokinetic-pharmacodynamic (PK-PD) methods. We have developed an in vitro PKPD model of TB which we have used to study resistance emergence to rifampin and isoniazid monotherapy during bactericidal activity. PK-PD studies of resistance emergence during sterilizing activity had as of yet not been performed. In the current application, we present PK-PD data on pyrazinamide sterilizing activity, setting the stage to study the PK-PD of combination anti-TB drugs during both bactericidal and sterilizing activity. Our long term goal is to develop anti-TB therapy dosing strategies that will lead to suppression of MDR-TB. The objective of the current application is to use the in vitro PK-PD model of TB to expose Mycobacterium tuberculosis to anti-TB drug pharmacokinetics similar to those encountered in humans. Our central hypothesis is that PK-PD optimization of combination anti-TB regimens will lead suppression of MDR-TB. We will test the hypothesis and achieve our objectives by examining four specific aims. (1) We will apply PK-PD methods to test the pharmacokinetic mismatch-, the effective monotherapy-, and the TMSW- hypothesis. (2) We will identify the exposures of first line anti-TB drug combination therapy that are associated with suppression of resistance during both bactericidal and sterilizing therapy. (3) We will identify the degree of poor compliance associated with emergence of resistance and identify PK-PD solutions to reduce emergence of MDR-TB, despite the poor compliance. We will also identify the order of emergence of chromosomal mutations that confer drug resistance during non-compliance. (4) We will translate the in vitro PK-PD exposures and strategies that suppress resistance to human therapy by use of Monte Carlo simulations. We will examine 2-drug and 3-drug combinations during both bactericidal and sterilizing activity in the in vitro model, and examine the size of resistant sub-populations that arise with each particular dosing strategy. Our in vitro model uses virulent M. tuberculosis and mimics the concentration-time profile of drugs encountered in TB patients. From the proposed studies, we expect some of the standard explanations on how resistance arises to be false, which will lead to fundamental changes in strategies to reduce MDR-TB. Since poor compliance is a fact of clinical life, dosing strategies that suppress drug resistance despite poor compliance would be a major therapeutic innovation. Tuberculosis is a global problem that has been declared a global emergence by the international community. A particularly big problem is the emergence of multidrug resistant tuberculosis. The current application will apply pharmacokinetic-pharmacodynamic solutions to this public health problem.